Article
Multidisciplinary Sciences
Miguel Ruiz-Garcia, Juan Ozaita, Maria Pereda, Antonio Alfonso, Pablo Branas-Garza, Jose A. Cuesta, Angel Sanchez
Summary: Networks of social interactions are crucial for civilizations, but the quantitative understanding of them is still limited. This study examines real social networks in 13 schools, involving over 3,000 students and 60,000 relationships. The triadic influence metric is introduced to measure the impact of nearest neighbors, which outperforms personal traits in predicting relationship signs.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2023)
Article
Computer Science, Artificial Intelligence
Shuo-shuo Zhang, Xiang-rong Tong, Shui-gen Wang
Summary: This article introduces an ELM-NeuralWalk algorithm for calculating indirect trust in online social networks. The algorithm learns trust calculation rules and performs iterative calculations to accurately calculate indirect trust between users.
INTERNATIONAL JOURNAL OF BIO-INSPIRED COMPUTATION
(2022)
Article
Engineering, Chemical
Jiajun Xu, Yuzhen Lu, Ebenezer Olaniyi, Lorin Harvey
Summary: This study presents a machine vision system for volume measurement of sweetpotatoes, using automated imaging and deep learning models to achieve accurate and non-destructive volume estimation.
JOURNAL OF FOOD ENGINEERING
(2024)
Article
Geography
Jingpeng Liao, Qiulin Liao, Weiwei Wang, Shouyun Shen, Yao Sun, Peng Xiao, Yuci Cao, Jiaao Chen
Summary: This study develops an intelligent landscape value evaluation method based on online texts, which converts online texts into quantities through deep learning technology to measure public preference, emotional tendency, and intensity. By collecting and analyzing online texts about Orange Island in Changsha city, the study shows that the method has a wide range of data sources and represents the real public perception and value orientation.
Article
Computer Science, Information Systems
Rachit Shukla, Adwitiya Sinha, Ankit Chaudhary
Summary: In the age of information super-connection, social media platforms have been misused by bots for spreading misinformation, manipulating public opinions, and promoting hidden agendas, posing a serious threat to social media. Therefore, we have proposed an AI-driven framework to identify Twitter bots and substantiated its effectiveness through research.
Article
Mathematics, Interdisciplinary Applications
Xuelian Ni, Fei Xiong, Shirui Pan, Hongshu Chen, Jia Wu, Liang Wang
Summary: In this study, the pattern of heterogeneous social influence on human decision-making from the perspectives of opinions, behaviors, preferences, and decision probabilities is investigated across three large-scale online networks. The results show that similar behavior during exploration has a stronger influence on decision-making compared to explicit social relationships, regardless of relationship sparsity, and is associated with shared experiences, network structures, and individual attributes.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Computer Science, Artificial Intelligence
Chenhui Guo, Xi Chen, Paulo Goes, Cheng Zhang
Summary: This study investigates the impact of social influence on users' spending time and spending money in a freemium model. The findings reveal that spending time has a stronger effect on spending money, and personal characteristics and network measures moderate this multiplex social influence.
DECISION SUPPORT SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yunpeng Ma, Chenheng Xu, Hua Wang, Shengkai Liu, Xiaoying Gu
Summary: This study proposes a novel method, SDLN, for time series online forecasting without capturing sliding time window features. By adding decomposition and sparse layers, SDLN achieves fast model calculation speed and better model accuracy and stability in a short time. Experimental results show that SDLN outperforms other state-of-the-art methods in terms of model computation speed and achieves high accuracy and fast computation time for short- and medium-term forecasting across different datasets.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Theory & Methods
Atika Mbarek, Salma Jamoussi, Abdelmajid Ben Hamadou
Summary: The use of social media platforms makes it easier for people to express their feelings and thoughts online, and active users on these platforms may show signals of suicidal ideation, highlighting the importance of detecting users with suicidal tendencies.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Psychology, Multidisciplinary
Jingfang Liu, Mengshi Shi
Summary: This study uses machine learning technology to detect users with depression by analyzing user-shared content and posting behaviors. A hybrid feature selection and stacking ensemble strategy is proposed to improve the recognition accuracy. The experimental results show that this method achieves a high accuracy of 90.27% in identifying online patients.
FRONTIERS IN PSYCHOLOGY
(2022)
Article
Education & Educational Research
R. A. Crane, S. Comley
Summary: Massive open online courses (MOOCs) are effective, flexible, and allow large participation in distance learning. Social Learners who engage in MOOC forums are more likely to complete the courses, despite being in the minority compared to Non-Social Learners. Further research is needed to understand the value of Social Learning and its impact on student retention in MOOCs.
EDUCATION AND INFORMATION TECHNOLOGIES
(2021)
Article
Mathematics
Lin Zhang, Kan Li
Summary: In this study, a novel framework called Influence Maximization based on Prediction and Replacement (IMPR) is proposed to address the influence maximization problem in dynamic online social networks. The framework utilizes historical network snapshot information to predict upcoming network snapshots, and employs a fast replacement algorithm to solve the seed node problem. Experimental results demonstrate the promising performance of the proposed scheme.
Article
Computer Science, Theory & Methods
Manuel F. Lopez-Vizcaino, Francisco J. Novoa, Victor Carneiro, Fidel Cacheda
Summary: The article discusses different approaches to early detection of cyberbullying on social networks, proposing two sets of features and two early detection methods that have successfully improved baseline detection models by up to 42%.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Mathematics
Lin Zhang, Kan Li
Summary: Influence maximization is a popular research issue in online social network analysis. We propose a new framework called Influence Maximization Based on Backward Reasoning, which achieves a balance between accuracy and efficiency while ensuring the accuracy of the algorithm.
Article
Computer Science, Information Systems
Yandi Li, Haobo Gao, Yunxuan Gao, Jianxiong Guo, Weili Wu
Summary: The article introduces the importance and challenges of Influence Maximization (IM) problem, compares traditional algorithms with ML-based methods, and summarizes the recent research progress. It emphasizes the advantages and application prospects of using machine learning methods such as Deep Reinforcement Learning to solve the IM problem, and points out the challenges that need to be addressed in future research.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2023)
Article
Computer Science, Information Systems
Qingyuan Gong, Yang Chen, Xinlei He, Yu Xiao, Pan Hui, Xin Wang, Xiaoming Fu
Summary: The study of social influence has gained much attention in academia, but identifying influential cold-start users on emerging social networks remains challenging. This research proposes a practical solution to predict whether a cold-start user will become influential by leveraging information from dominant social networks. The system demonstrates high prediction performance based on different social influence definitions after being evaluated with real data collected from dominant and emerging social networks.
ACM TRANSACTIONS ON THE WEB
(2021)
Article
Computer Science, Software Engineering
Truong An Pham, Junjue Wang, Roger Iyengar, Yu Xiao, Padmanabhan Pillai, Roberta Klatzky, Mahadev Satyanarayanan
Summary: WCA enhances human cognition in real time through wearable devices and edge computing infrastructure. It has the potential to significantly impact various fields, but current development challenges include high skill requirements and slow progress.
SOFTWARE-PRACTICE & EXPERIENCE
(2021)
Article
Computer Science, Information Systems
Hui Gao, Jianhao Feng, Yu Xiao, Bo Zhang, Wendong Wang
Summary: In this paper, a UAV-assisted MCS method is proposed to optimize the sensing coverage and data quality. The method uses deep reinforcement learning to schedule UAV trajectories and sensing activities, minimizing overall energy cost. Simulation results show that the proposed scheme outperforms compared methods in terms of coverage completed ratio, calibrating ratio, energy efficiency, and task fairness.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Computer Science, Information Systems
Wencan Mao, Ozgur Umut Akgul, Abbas Mehrabi, Byungjin Cho, Yu Xiao, Antti Yla-Jaaski
Summary: This study focuses on the capacity planning problem in vehicular fog computing (VFC) and proposes a data-driven framework to optimize the deployment of fog nodes for cost minimization. The results show that high traffic density and hourly variations lead to more deployment of mobile fog nodes and cost savings.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Computer Science, Hardware & Architecture
Clayton Frederick Souza Leite, Yu Xiao
Summary: This paper proposes a resource-efficient and high-performance continual learning solution for sensor-based human activity recognition (HAR). By utilizing a neural network trained with a replay-based method and a highly-compressed replay memory, the method achieves accuracy improvements and faster operation on low-cost resource-constrained devices.
ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Petr Byvshev, Pascal Mettes, Yu Xiao
Summary: This work investigates the bias towards static appearance in 3D convolutional networks and questions its relationship with the limited significance of motion in training data. By introducing temporality measures and synthetic datasets, the study reveals that 3D architectures are not inherently biased towards appearance and that the bias arises from training on specific video sets. Moreover, the proposed measures provide a reliable method to estimate motion relevance and uncover differences between popular pre-training video collections.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2022)
Article
Materials Science, Ceramics
Zhang Bo, Lixia Zhang, Pan Hui, Sun Zhan, Chang Qing
Summary: A reliable diffusion bonding of Ti3Si(Al)C-2 ceramic is achieved by using Au foil as an interlayer at a lower temperature and shorter time. This method significantly decreases the bonding temperature by about 150 degrees Celsius compared to the lowest temperature reported in current research, and achieves a higher shear strength at 650 degrees Celsius.
JOURNAL OF THE EUROPEAN CERAMIC SOCIETY
(2022)
Article
Computer Science, Artificial Intelligence
Qingyuan Gong, Yushan Liu, Jiayun Zhang, Yang Chen, Qi Li, Yu Xiao, Xin Wang, Pan Hui
Summary: In this work, a deep learning-based solution called GitSec is proposed for detecting malicious accounts in online developer communities. By analyzing account profiles, dynamic activity characteristics, and social interactions, GitSec successfully distinguishes malicious accounts from legitimate ones, and comprehensive evaluations show that it outperforms existing solutions.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Theory & Methods
Xuebing Li, Yang Chen, Mengying Zhou, Tiancheng Guo, Chenhao Wang, Yu Xiao, Junjie Wan, Xin Wang
Summary: This article presents Artemis, a low-latency naming and routing system that improves internet service deployment by optimizing server selection and reducing query latency.
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Yuwei Zhang, Qingyuan Gong, Yang Chen, Yu Xiao, Xin Wang, Pan Hui, Xiaoming Fu
Summary: Location-Based Services (LBS) have thrived due to the technological advancements in smart devices. Analyzing user-generated data based on location is a useful method to understand human mobility patterns, and it has also contributed to the development of applications such as recommender systems and urban computing. This dataset from LBSLab, a smartphone-based system within the WeChat app, documents user activities including logins, profile viewings, weather checks, and check-ins with location information, collected from 467 users over an 11-day period. The dataset also provides temporal and spatial data analysis, offering insights into user behaviors, human mobility, and the temporal and spatial characteristics of people's moods.
Article
Computer Science, Information Systems
Ozgur Umut Akgul, Wencan Mao, Byungjin Cho, Yu Xiao
Summary: Edge/fog computing plays a crucial role in meeting the low latency requirements of emerging vehicle applications in 5G and beyond, such as autonomous driving. Vehicular fog computing (VFC) is a cost-effective deployment option that combines stationary fog nodes with mobile ones carried by vehicles. To effectively plan and manage VFC resources, it is important to consider spatiotemporal variations in demand and supply, tradeoffs between quality-of-service and costs, and real-world scenarios. Existing simulators lack realism, mobility support for fog nodes, and realistic testing scenarios. In this article, we propose VFogSim, an open-source simulator that enables simulation of VFC supply and demand using real-world data in urban areas. It supports various vehicular traffic models, evaluates deployment scenarios, and validates accuracy using real-world measurements.
IEEE SYSTEMS JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Byungjin Cho, Yu Xiao
Summary: The study focuses on decentralized decision-making in vehicular fog computing scenarios, where each agent aims to minimize their costs by offloading tasks to specific VFNs. They utilize a completely uncoupled learning rule to adapt to dynamicity and uncertainty in the system.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Aziza Zhanabatyrova, Clayton Frederick Souza Leite, Yu Xiao
Summary: Autonomous driving requires accurate and up-to-date 3-D maps for semantic landmarks. Vision-based mapping solutions using camera data have attracted attention, but automatic change detection remains an open issue. We propose a pipeline for initiating and updating 3-D maps with dashcam videos, focusing on automatic change detection based on metadata comparison. We introduce a deep learning-based 3-D localization algorithm and a point clustering algorithm to improve the system's performance.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Proceedings Paper
Engineering, Electrical & Electronic
Xinlei Xie, Ruoyi Zhang, Chao Zhu, Ruijin Li, Xiangyuan Bu, Yu Xiao
Summary: This paper proposes a multi-task allocation strategy (EAAV) based on antenna array for simultaneously offloading multiple tasks in vehicular fog computing. The strategy aims to reduce transmission power consumption while maintaining high transmission data rate, considering the mobility of vehicles and communication interference.
2022 IEEE 95TH VEHICULAR TECHNOLOGY CONFERENCE (VTC2022-SPRING)
(2022)
Article
Computer Science, Information Systems
Truong An Pham, Tim Moesgen, Sanni Siltanen, Joanna Bergstrom, Yu Xiao
Summary: Annotated videos are commonly used in the manufacturing industry to document assembly and maintenance processes, but the current process is cumbersome and time-consuming. To address this, ARiana, an augmented reality-based video annotation tool, has been developed to efficiently create annotations while conducting assembly or maintenance tasks.
